Curated by THEOUTPOST
On Sun, 16 Mar, 4:00 PM UTC
2 Sources
[1]
Researchers develop computational tools to safeguard privacy without degrading voice-based cognitive markers
However, voice data introduces privacy challenges due to the personally identifiable information embedded in recordings, such as gender, accent and emotional state, as well as more subtle speech characteristics that can uniquely identify individuals. These risks are amplified when voice data is processed by automated systems, raising concerns about re-identification and potential misuse of data. In a new study, researchers from Boston University Chobanian & Avedisian School of Medicine have introduced a computational framework that applies pitch-shifting, a sound recording technique that changes the pitch of a sound, either raising or lowering it, to protect speaker identity while preserving acoustic features essential for cognitive assessment. "By leveraging techniques such as pitch-shifting as a means of voice obfuscation, we demonstrated the ability to mitigate privacy risks while preserving the diagnostic value of acoustic features," explained corresponding author Vijaya B. Kolachalama, PhD, FAHA, associate professor of medicine. Using data from the Framingham Heart Study (FHS) and DementiaBank Delaware (DBD), the researchers applied pitch-shifting at different levels and incorporated additional transformations, such as time-scale modifications and noise addition, to alter vocal characteristics to responses to neuropsychological tests. They then assessed speaker obfuscation via equal error rate and diagnostic utility through the classification accuracy of machine learning models distinguishing cognitive states: normal cognition (NC), mild cognitive impairment (MCI) and dementia (DE). Using obfuscated speech files, the computational framework was able to accurately determine NC, MCI and DE differentiation in 62% of the FHS dataset and 63% of the DBD dataset. According to the researchers, this work contributes to the ethical and practical integration of voice data in medical analyses, emphasizing the importance of protecting patient privacy while maintaining the integrity of cognitive health assessments. "These findings pave the way for developing standardized, privacy-centric guidelines for future applications of voice-based assessments in clinical and research settings," adds Kolachalama, who also is an associate professor of computer science, affiliate faculty of Hariri Institute for Computing and a founding member of the Faculty of Computing & Data Sciences at Boston University. These findings appear online in Alzheimer's & Dementia: The Journal of the Alzheimer's Association. This project was supported by grants from the National Institute on Aging's Artificial Intelligence and Technology Collaboratories (P30-AG073104 and P30-AG073105), the American Heart Association (20SFRN35460031), Gates Ventures, and the National Institutes of Health (R01-HL159620, R01-AG062109, and R01-AG083735).
[2]
Computational framework protects privacy in voice-based cognitive health assessments
Boston University School of MedicineMar 14 2025 Digital voice recordings contain valuable information that can indicate an individual's cognitive health, offering a non-invasive and efficient method for assessment. Research has demonstrated that digital voice measures can detect early signs of cognitive decline by analyzing features such as speech rate, articulation, pitch variation and pauses, which may signal cognitive impairment when deviating from normative patterns. However, voice data introduces privacy challenges due to the personally identifiable information embedded in recordings, such as gender, accent and emotional state, as well as more subtle speech characteristics that can uniquely identify individuals. These risks are amplified when voice data is processed by automated systems, raising concerns about re-identification and potential misuse of data. In a new study, researchers from Boston University Chobanian & Avedisian School of Medicine have introduced a computational framework that applies pitch-shifting, a sound recording technique that changes the pitch of a sound, either raising or lowering it, to protect speaker identity while preserving acoustic features essential for cognitive assessment. "By leveraging techniques such as pitch-shifting as a means of voice obfuscation, we demonstrated the ability to mitigate privacy risks while preserving the diagnostic value of acoustic features," explained corresponding author Vijaya B. Kolachalama, PhD, FAHA, associate professor of medicine. Using data from the Framingham Heart Study (FHS) and DementiaBank Delaware (DBD), the researchers applied pitch-shifting at different levels and incorporated additional transformations, such as time-scale modifications and noise addition, to alter vocal characteristics to responses to neuropsychological tests. They then assessed speaker obfuscation via equal error rate and diagnostic utility through the classification accuracy of machine learning models distinguishing cognitive states: normal cognition (NC), mild cognitive impairment (MCI) and dementia (DE). Using obfuscated speech files, the computational framework was able to accurately determine NC, MCI and DE differentiation in 62% of the FHS dataset and 63% of the DBD dataset. According to the researchers, this work contributes to the ethical and practical integration of voice data in medical analyses, emphasizing the importance of protecting patient privacy while maintaining the integrity of cognitive health assessments. "These findings pave the way for developing standardized, privacy-centric guidelines for future applications of voice-based assessments in clinical and research settings," adds Kolachalama, who also is an associate professor of computer science, affiliate faculty of Hariri Institute for Computing and a founding member of the Faculty of Computing & Data Sciences at Boston University. These findings appear online in Alzheimer's & Dementia: The Journal of the Alzheimer's Association. This project was supported by grants from the National Institute on Aging's Artificial Intelligence and Technology Collaboratories (P30-AG073104 and P30-AG073105), the American Heart Association (20SFRN35460031), Gates Ventures, and the National Institutes of Health (R01-HL159620, R01-AG062109, and R01-AG083735). Boston University School of Medicine Journal reference: Ahangaran, M., et al. (2025). Obfuscation via pitch‐shifting for balancing privacy and diagnostic utility in voice‐based cognitive assessment. Alzheimer's & Dementia. doi.org/10.1002/alz.70032.
Share
Share
Copy Link
Researchers at Boston University have developed a computational framework using AI techniques to protect privacy in voice-based cognitive health assessments, balancing data security with diagnostic accuracy.
Researchers from Boston University Chobanian & Avedisian School of Medicine have developed a groundbreaking computational framework that addresses the critical challenge of maintaining privacy in voice-based cognitive health assessments. This innovative approach utilizes artificial intelligence techniques to protect speaker identity while preserving the diagnostic value of acoustic features essential for cognitive evaluation 12.
Digital voice recordings have emerged as a valuable tool for assessing cognitive health, offering a non-invasive and efficient method for detecting early signs of cognitive decline. By analyzing features such as speech rate, articulation, pitch variation, and pauses, researchers can identify deviations from normative patterns that may indicate cognitive impairment 2.
However, the use of voice data in healthcare introduces significant privacy concerns. Voice recordings contain personally identifiable information, including gender, accent, emotional state, and subtle speech characteristics that can uniquely identify individuals. These privacy risks are further amplified when voice data is processed by automated systems, raising concerns about potential re-identification and misuse of sensitive health information 12.
To address these privacy challenges, the Boston University team has introduced a computational framework that leverages pitch-shifting, a sound recording technique that alters the pitch of a sound. This approach aims to protect speaker identity while maintaining the acoustic features crucial for cognitive assessment 1.
Dr. Vijaya B. Kolachalama, the study's corresponding author, explained, "By leveraging techniques such as pitch-shifting as a means of voice obfuscation, we demonstrated the ability to mitigate privacy risks while preserving the diagnostic value of acoustic features" 12.
The researchers applied their framework to data from the Framingham Heart Study (FHS) and DementiaBank Delaware (DBD), implementing pitch-shifting at various levels and incorporating additional transformations such as time-scale modifications and noise addition. These techniques were used to alter vocal characteristics in responses to neuropsychological tests 12.
The effectiveness of the framework was assessed through:
The models were tasked with distinguishing between three cognitive states: normal cognition (NC), mild cognitive impairment (MCI), and dementia (DE) 12.
Using the obfuscated speech files, the computational framework achieved impressive results:
This groundbreaking work contributes significantly to the ethical and practical integration of voice data in medical analyses. It emphasizes the critical importance of protecting patient privacy while maintaining the integrity of cognitive health assessments 12.
Dr. Kolachalama, who also holds positions as an associate professor of computer science and is affiliated with the Hariri Institute for Computing at Boston University, added, "These findings pave the way for developing standardized, privacy-centric guidelines for future applications of voice-based assessments in clinical and research settings" 12.
The study's findings have been published in Alzheimer's & Dementia: The Journal of the Alzheimer's Association, marking a significant step forward in the field of AI-powered healthcare privacy protection 12.
A University at Buffalo-led study introduces a novel encryption technique for AI-powered medical data, proving highly effective in detecting sleep apnea while safeguarding patient privacy.
2 Sources
2 Sources
Researchers from Iraq and Australia highlight the potential of AI algorithms to detect Parkinson's disease through subtle voice changes, offering a new approach for early diagnosis and remote monitoring.
3 Sources
3 Sources
Researchers at Klick Labs have developed a groundbreaking voice analysis technology that can detect hypertension with high accuracy. This non-invasive method could revolutionize blood pressure screening and monitoring.
2 Sources
2 Sources
Researchers develop a brain-computer interface that can translate thoughts into audible speech almost instantly, potentially revolutionizing communication for people with severe paralysis.
18 Sources
18 Sources
Researchers have developed an AI-based voice analysis technology that can detect undiagnosed type 2 diabetes. This innovative approach could revolutionize early screening and diagnosis of the disease.
2 Sources
2 Sources
The Outpost is a comprehensive collection of curated artificial intelligence software tools that cater to the needs of small business owners, bloggers, artists, musicians, entrepreneurs, marketers, writers, and researchers.
© 2025 TheOutpost.AI All rights reserved